| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4830917874396135, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00966183574879227, | |
| "eval_loss": 10.882368087768555, | |
| "eval_runtime": 2.9101, | |
| "eval_samples_per_second": 239.851, | |
| "eval_steps_per_second": 30.239, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0966183574879227, | |
| "grad_norm": 18.205909729003906, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 10.4561, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0966183574879227, | |
| "eval_loss": 9.441638946533203, | |
| "eval_runtime": 2.8968, | |
| "eval_samples_per_second": 240.952, | |
| "eval_steps_per_second": 30.378, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1932367149758454, | |
| "grad_norm": 11.83459186553955, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 8.7933, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1932367149758454, | |
| "eval_loss": 8.16073226928711, | |
| "eval_runtime": 2.9153, | |
| "eval_samples_per_second": 239.423, | |
| "eval_steps_per_second": 30.185, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2898550724637681, | |
| "grad_norm": 9.983260154724121, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 7.8034, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2898550724637681, | |
| "eval_loss": 7.373321056365967, | |
| "eval_runtime": 2.9407, | |
| "eval_samples_per_second": 237.36, | |
| "eval_steps_per_second": 29.925, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3864734299516908, | |
| "grad_norm": 9.781479835510254, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 7.1231, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3864734299516908, | |
| "eval_loss": 7.0239973068237305, | |
| "eval_runtime": 3.0092, | |
| "eval_samples_per_second": 231.954, | |
| "eval_steps_per_second": 29.244, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4830917874396135, | |
| "grad_norm": 9.775260925292969, | |
| "learning_rate": 0.0, | |
| "loss": 6.9844, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4830917874396135, | |
| "eval_loss": 6.96579647064209, | |
| "eval_runtime": 2.9634, | |
| "eval_samples_per_second": 235.542, | |
| "eval_steps_per_second": 29.696, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 91781962137600.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |